Differences between completers and early dropouts from 2 HIV intervention trials: a health belief approach to understanding prevention program attrition.

OBJECTIVES: The purpose of this study was to identify factors predicting program attrition among participants in human immunodeficiency virus (HIV) risk reduction trials.
METHODS: Participants were gay/bisexual men and severely mentally ill adults recruited to take part in HIV risk reduction small-group interventions. Program completers were compared with participants who were assessed at baseline but then failed to attend any sessions. The health belief model provided a framework for selection of possible predictors of program attrition.
RESULTS: Younger age was associated with early dropout in both samples. Other predictors among gay/bisexual men included involvement in an exclusive sexual relationship, minority ethnicity, injection drug use, and higher perceived severity of AIDS. Severely mentally ill dropouts were less knowledgeable about safer sex methods and more likely to hold positive outcome expectancies for condom use.
CONCLUSIONS: Evaluation of intervention effectiveness among vulnerable population segments is threatened if there is selective attrition. Better methods are needed to attract and maintain participation in HIV prevention programs. Alternatively, wider application of "intention to treat" analysis of intervention outcomes is recommended to minimize selection bias due to program dropout.